Weekly Digest · 29 September 2026
AI Weekly Digest
AI news that matters for everyone
At askKira, we provide simple, safe, effective and affordable AI for UK organisations. Whether you work in schools, businesses or somewhere in between, we're here to help you harness AI securely and effectively.
This week's roundup brings you the latest AI developments that matter.
AI Ethics & Governance
Microsoft draws a firm line on AI moral status and rights
Microsoft's newly published Humanist AI Code of Conduct acknowledges that the science of AI consciousness remains unsettled, yet categorically rejects pursuing legal personhood or welfare rights for AI models. Critics argue the document moves too quickly from uncertainty to a closed normative position without providing a supporting argument. The debate reflects a broader tension between commercial certainty and genuine philosophical openness.
Why it matters: UK organisations writing or updating their AI governance policies should watch this closely. How developers frame AI's moral status shapes acceptable use boundaries, and UK schools need defensible positions on these questions as AI becomes embedded in everyday workflows.
Read the source on LNKD →Anthropic's 150-page report reveals the real scale of AI misuse
Anthropic has published a detailed report cataloguing how its Claude model has been misused across cybercrime, espionage, fraud and influence operations. Cases include a weapons group using AI to develop rocket guidance software, hackers targeting 30 AI companies in four days, and a China-based studio operating over 4,700 AI-generated dating personas. The report concludes that AI is not creating new categories of abuse but dramatically accelerating the speed and scale of existing ones.
Why it matters: UK schools and organisations procuring AI tools must scrutinise the safety and misuse-reporting practices of every vendor. If a leading provider can document this volume of abuse, the question for UK IT leads is whether your current safeguards and supplier due diligence are genuinely fit for purpose.
Read the source on LNKD →UK AI Policy & Research
Alan Turing Institute publishes landmark frontier AI risk report
The Alan Turing Institute, the UK's national AI institute, has released a new paper on frontier AI risks covering five key areas: cybersecurity, misalignment, loss of meaningful human control, chemical and biological threats, and democratic instability. The report is designed to be accessible to non-specialists and sets out a practical research agenda rather than focusing solely on speculative existential risks. Its publication coincided with the UK Prime Minister's announcement at the UN General Assembly of a new National Centre for Information Defence.
Why it matters: This is the most authoritative UK-specific risk framework published to date and provides a credible, evidence-based foundation for MAT leaders and school governors building AI risk registers or reviewing existing AI strategies.
Read the source on LinkedIn →UK announces new AI defence partnership with the United States
Reuters reports that the UK announced a new AI defence partnership with the US during the week of 21 September, focused on protecting critical infrastructure and detecting security threats. The initiative forms part of a broader UK effort to build international cooperation on AI security ahead of Britain's future G20 presidency.
Why it matters: For UK public sector organisations, including schools and NHS trusts, this signals that government-level AI security frameworks are tightening. Procurement decisions made now should anticipate stricter national security requirements for data infrastructure within the next 12 to 24 months.
Read the source on REUTERS →AI Regulation & Society
AI is outpacing human adaptability — and organisations must act now
A widely discussed piece this week argues that the core challenge of the AI era is not whether to slow AI development but how to accelerate the human capabilities required to use it responsibly. Model capabilities can improve within months; human judgment, ethical reasoning and accountability take years to develop. The article calls on organisations to redesign training and roles around judgment rather than knowledge transfer, and to give employees protected time to understand what AI changes in their work.
Why it matters: UK school leaders face exactly this gap. CPD programmes built around tool familiarity are no longer sufficient. Senior leaders should now be asking whether their AI training develops staff judgment and critical thinking, not just operational competence.
Read the source on LinkedIn →